In a surprising move, the head of Instagram, Adam Mosseri, shared in a blog post how the Instagram algorithm works.
In the post, he wants to answer questions such as: “How does Instagram decide what comes up first for me?”; “Why do some of my posts get more views than others?”; and “How does Instagram decide what to show me in Explore?”
Mosseri says the main “misunderstanding” is about the Instagram algorithm, which is actually “a series of algorithms, classifiers and processes, each with its own purpose.”
Each part of the app ̵1; Feed, Explore, Reels – uses its own algorithm tailored to how people use it. People tend to look for their closest friends in Stories, but they want to discover something completely new in Explore. We rank things differently in different parts of the app, based on how people use them.
About Instagram Feed and Stories says the leader of the app Instagram focuses on posts from people you follow. With the information about what was posted, the people who made these posts, and the user’s preference, the app finds “signals”. These are the signals that Instagram uses in order of importance:
- Information about the post. These are signals both about how popular a post is – think how many people have liked it – and more worldly information about the content itself, such as when it was posted, how long it is if it is a video, and what location, if any , was attached to it.
- Information about the person who posted. This helps us get a sense of how interesting the person may be to you, and includes cues such as how many times people have communicated with that person in recent weeks.
- Your activity. This helps us understand what you might be interested in and includes signals such as how many posts you liked.
- Your story of interacting with someone. This gives us a sense of how interested you are in general in seeing posts from a particular person. An example is whether you comment on each other’s posts or not.
Adam Mosseri says Instagram is trying to avoid showing too many posts from the same person in a row. Another example is about stories that were forwarded from Feed:
Until recently, we valued these stories less, because we have consistently heard that people are more interested in seeing original stories. But we see a swarm of forwarded posts in big moments – everything from the World Cup to social unrest – and in those moments people expected their stories to reach more than they did, so we stopped.
In the post, Mosseri also explains what Instagram uses to rank posts on the Explore page and in the wheel section. To improve the “training” algorithms, there are some tips:
- Choose your close friends;
- Mute people you are not interested in;
- Mark recommended posts as “Not interested.”
FTC: We use auto affiliate links. More.
Check out 9to5Mac on YouTube for more Apple news: